A flatbed type image reading device has a read section for reading a medium to be read. When the medium to be read is set on the read section, the read section reads the medium to be read. In order for the image reading device to be provided as a product, the image reading device has to have features of low price and of portability. Therefore, a size of the read section provided in the image reading device is restricted. However, when such a small-sized read section is used, a size of the medium to be read is also limited, for example, a whole image of a large size sheet cannot be obtained. Accordingly, an image combining technology has been developed, so that a medium to be read having a larger size than the read section can be read. For this combining technology, a highly precise image combining process is required.
Japanese Patent Application Laid-open Publication No. 6-36009 discloses an image editing device. The image editing device includes image reading means, image editing means, storage means for storing an image after image editing and combining means. The image reading means reads an image of an original being within a predetermined read range, and outputs an image signal. The image editing means stores the read image, sets plural reference points on the image, and performs an image rotation process, an image enlarging process and an image reducing process. The combining means combines these images by referring to the reference points set on the image stored in the storage means and corresponding reference points set on each of the images, which are newly stored. When a size of an original to be read is larger than the size of the predetermined read range of the image editing device, the image editing device virtually divides the image of the original into smaller blocks, and the image reading means reads each of the virtually divided images of the original. The partial images read as above are combined each other by the storage means according to the reference points set on each of the partial images. As a result, an image of the original, a size of which is larger than the size of the predetermined reading range is stored in the storage means.
In the image editing device according to the related art, the reference points used for combining images is set after reading and storing the virtually divided images according to a mark which is set to the original before reading the original in advance.
Japanese Patent Application Laid-open Publication No. 11-88655 discloses an image processing apparatus in which plural images are separately inputted, and are joined. A single image is generated from the plural divided images. The image processing apparatus includes image input means for inputting an image, image storing means for storing data of the plural divided images which are outputted by the image input means, joint position determination means for determining a joint position for each scan line in the plural images and joint process means for performing a joint process to the plural images according to the joint position determined by the joint position determination means. In the image processing method in the related art, a correlation is evaluated for each pixel in the image areas that are read with respect to the images to be joined, and the joint position is determined.
Japanese Patent Application Laid-open Publication No. 2000-188675 discloses an image division input method used for an image input device in which an original on a platen is scanned and the image of the original is read. The image of the original whose size is larger than a size of a scan range is virtually divided into small blocks and each of the virtually divided images is read. In the image division input method of the related art, two scanning processes, i.e. a coarse scan and a fine scan, are performed. In a first coarse scan image obtained by the coarse scan, an area where a fine scan is performed and a connection area adjacent to the fine scan area are set. A second coarse scan area is obtained by performing the coarse scan in the area adjacent to the connection area. The fine scan area is definitively set by adjusting the position of the second coarse scan area and the position of the connection area. The combined image is obtained by repeating the above mentioned processes.